COSMOGRAIL: the COSmological MOnitoring of GRAvItational Lenses ^⋆,⋆⋆

XIII. Time delays and 9-yr optical monitoring of the lensed quasar RX J1131−1231

¹ Laboratoire d’astrophysique, École Polytechnique Fédérale de Lausanne (EPFL), Observatoire de Sauverny, 1290 Versoix, Switzerland
e-mail: malte.tewes@epfl.ch
² Department of Astronomy, The Ohio State University, 140 West 18th Av., Columbus, OH 43210, USA
³ Center for Cosmology and Astroparticle Physics, The Ohio State University, 191 West Woodruff Av., Columbus, OH 43210, USA
⁴ Institut d’Astrophysique et de Géophysique, Université de Liège, Allée du 6 Août, 17, 4000 Sart Tilman, Liège 1, Belgium
⁵ Instituut voor Sterrenkunde, Katholieke Universiteit Leuven, Celestijnenlaan 200B, 3001 Heverlee, Belgium
⁶ Argelander-Institut für Astronomie, Auf dem Hügel 71, 53121 Bonn, Germany
⁷ School of Physics and Astronomy, University of Nottingham, University Park, Nottingham NG7 2RD, UK

Received: 8 September 2012
Accepted: 31 May 2013

Abstract

We present the results from nine years of optically monitoring the gravitationally lensed z_QSO = 0.658 quasar RX J1131−1231. The R-band light curves of the four individual images of the quasar were obtained using deconvolution photometry for a total of 707 epochs. Several sharp quasar variability features strongly constrain the time delays between the quasar images. Using three different numerical techniques, we measured these delays for all possible pairs of quasar images while always processing the four light curves simultaneously. For all three methods, the delays between the three close images A, B, and C are compatible with being 0, while we measured the delay of image D to be 91 days, with a fractional uncertainty of 1.5% (1σ), including systematic errors. Our analysis of random and systematic errors accounts in a realistic way for the observed quasar variability, fluctuating microlensing magnification over a broad range of temporal scales, noise properties, and seasonal gaps. Finally, we find that our time-delay measurement methods yield compatible results when applied to subsets of the data.